What causes engine roughness when you lean too much?

[MountainDude]

O-470S on a C182, though I think this applies to most engines.
When I lean too much, the engine starts running rough. What is actually happening at that time? I thought as I lean, the leanest cylinder may not have enough fuel to combust, which would cause EGTs to plummet, but that does not happen. It looks like when it runs rough, all cylinders are combusting, so why is it rough?

 

[GoFly]

Leaning too much is how you turn off the engine when your flight is over.

 

[MountainDude]

Leaning too much is how you turn off the engine when your flight is over.

Sorry, I was not talking about that. I am referring to leaning in cruise.

 

[455 Bravo Uniform]

When you get too lean (or too rich on the other end) you begin to affect the ignitability. The magnetos and spark plugs can’t do their thing as effectively. Too much air, not enough fuel. Think misfire every so often.

As well, if the mixtures in each cylinder are uneven, your power pulses will be of different strengths, thus rough.

That’s a simple explanation.

 

[MountainDude]

When you get too lean (or too rich on the other end) you begin to affect the ignitability. The magnetos and spark plugs can’t do their thing as effectively. Too much air, not enough fuel. Think misfire every so often.

As well, if the mixtures in each cylinder are uneven, your power pulses will be of different strengths, thus rough.

That’s a simple explanation.

That makes sense. Thank you.

 

[mandm]

O-470S on a C182, though I think this applies to most engines.
When I lean too much, the engine starts running rough. What is actually happening at that time? I thought as I lean, the leanest cylinder may not have enough fuel to combust, which would cause EGTs to plummet, but that does not happen. It looks like when it runs rough, all cylinders are combusting, so why is it rough?

Fuel also acts as a coolant. Less fuel = hotter temps

 

[Hang 4]

Fuel also acts as a coolant. Less fuel = hotter temps

Until you get lean of peak, then less fuel=cooler temps.
This thread will now reach 30 pages

 

[William Pete Hodges]

The real culprit is uneven air fuel ratios between the cylinders. As you lean to the Lean of Peak EGT side of the graph, the power starts to drop off. If the air fuel ratio is the same in every cylinder, then every cylinder produces the same amount of power and the engine runs smoothly as you lean. But aviation engine don't always act like that, especially not carburerated flat sixes. They tend to muck up the fuel distribution and some of the cylinders run leaner while others run richer. When this happens each cylinder is producing a different amount of power that results in the engine running rough. Look at the power curve in the best economy band. It varies from 95% down to 85%, but in the best power band the curve it doesn't vary hardly at all. That means if your engine is running rough LOP then as you get richer the engine runs smooth again as the power per cylinder evens out.

This can also happen in flat fours but not to the same degree as flat sixes. Due to larger intake stroke overlap on six cylinder engines the mixture in the intake runners can partially reverse direction under certain conditions, making fuel distribution discrepancies worse. Fuel injected engines can often correct this with GAMI injectors that are adjusted to match their outputs to each cylinder.

Carburerated engines may have a tougher time with this but there are few tricks that may help. First, back off the throttle enough to induce turbulence into the mixture as it passes the throttle plate. This mixes the fuel and air into a more homogeneous mixture before splitting off into the intake runners. If this doesn't work by itself try adding some carb heat to help vaporize the fuel better. These two things may allow the engine to run leaner than before. None of this is guaranteed, but it should help. Try it at a high altitude cruise at about 60% power. You may find you can run leaner than you could before.

Hope this helps.

 

[midlifeflyer]

Fuel also acts as a coolant. Less fuel = hotter temps

Until you get lean of peak, then less fuel=cooler temps.
This thread will now reach 30 pages

You are correct but I think you may be talking apples and oranges since both ROP and LOP result in cooler CHT temperatures than within the "red box" (or "fin" if you prefer). And fuel flowing also results in overall cooling, which is why enriching is the usual response to abnormally high oil temps during climbout on a hot day.

 

[midlifeflyer]

Sorry, I was not talking about that. I am referring to leaning in cruise.

Isn't it ultimately the same thing at work?

 

[Capt. Geoffrey Thorpe]

All kinds of fun stuff starts happening when you get too lean. The air/fuel mixture is harder to ignite, and burns slower and less consistently. So, if you are sufficiently lean, you can get complete misfires, but for the initial roughness, it is more that the mixture is only partly burned during the combustion stroke. This leaves some unburned fuel/air in the cylinder so the next engine cycle there is less residual burned gas and the next cycle is easier to ignite and gives a bit more power. This significantly contributes to roughness. Variations in mixture between cylinders also contributes to roughness due to the differences in torque provided by each cylinder.

Temperatures tend to go down because there is less heat being generated (less fuel being burned). Slow burns can increase temperatures in the exhaust if the combustion continues there, but that is more of a problem for cars with catalysts than aircraft.

Power is down, fuel efficiency goes down because of the late nature of the burn and, if you lean enough, unburned hydrocarbons being pushed out the exhaust.

 

[William Pete Hodges]

You are correct but I think you may be talking apples and oranges since both ROP and LOP result in cooler CHT temperatures than within the "red box" (or "fin" if you prefer). And fuel flowing also results in overall cooling, which is why enriching is the usual response to abnormally high oil temps during climbout on a hot day.

There was nothing in his question about cooling nor did my answer refer to it. Cooler CHTs have more to do with the speed of combustion rather than the cooling affect of richer fuel. As air fuel ratio changes it has a dramatic effect on flame propagation time. When the mixture is most powerful ROP, flame propagation is so fast the peak cylinder pressures occur too close to TDC. This causes engine stress to rise because the pressure has no leverage to push the crank causing inter pressure to multiply and CHT to rise. From here, leaning or richening the mixture slows down flame propagation, allowing peak pressure to occur at a better angle to push the crank, producing more power with less internal pressure and stress.

 

[midlifeflyer]

There was nothing in his question about cooling nor did my answer refer to it.

Whatever makes you think that a response to posts by Hang 4 and mandm (even quoting them) had anything to do with your answer to the OP?

 

[William Pete Hodges]

Whatever makes you think that a response to posts by Hang 4 and mandm (even quoting them) had anything to do with your answer to the OP?

I am sorry if there has been a misunderstanding here. My original post was in reply to the OP. My second post was only a reply to you as you brought up other issues I had not replied to. Those are my only two posts. To the best of my knowledge I only quoted you.

 

[midlifeflyer]

I am sorry if there has been a misunderstanding here. My original post was in reply to the OP. My second post was only a reply to you as you brought up other issues I had not replied to. Those are my only two posts. To the best of my knowledge I only quoted you.

Yeah, but the point is, I was quoting others and not disagreeing with anything you said.

 

[MBDiagMan]

Well, since leaning is providing less fuel to the engine, maybe that has something to do with the engine running rough.

 

[Hang 4]

You are correct but I think you may be talking apples and oranges since both ROP and LOP result in cooler CHT temperatures than within the "red box" (or "fin" if you prefer). And fuel flowing also results in overall cooling, which is why enriching is the usual response to abnormally high oil temps during climbout on a hot day.

OP never talked oil temp. I was merely responding to the more fuel=cooler. (CHT and EGT), same is true leaner. Not sure how that's different. Agree that leaning in climb if things get hot isn't a great idea. But also looking at CHT, not oil temp.

 

[Dan Thomas]

One overlooked factor is spark. Weak sparkplugs will fail to ignite a lean mixture sooner that good plugs, so a plug that is dropping out will create roughness as one cylinder's burn is slower.

 

[Martin Pauly]

All kinds of fun stuff starts happening when you get too lean. The air/fuel mixture is harder to ignite

Correct. Which is a good reason to check the mags in cruise at a LOP power setting, e.g. at the top of descent. You will likely find an ignition issue earlier that way, early enough to fly somewhere where maintenance can be done conveniently.

A bit off topic, I know...

- Martin

 

[flyingron]

Fuel also acts as a coolant. Less fuel = hotter temps

Less combustion also means hotter temps. Temps peak at the stoichiometric ideal (all fuel and air used up). Too much of either one cools things down.

The roughness you get from excessive leaning is just the cylinder not firing due to not having enough fuel. As long as you are at cruise settings, this is harmless. At high power settings it may leave fuel in the cylinder that gets too rich the next cycle (similar to your click-click-click-bang! mag check).

Detonation actually occurs on the just rich side. That is bad at high power settings.

However Continental says, as long as you are at or below the max continuous, you can run anything from full rich to so lean the engine is misfiring without hurting anything.

 

[MountainDude]

The real culprit is uneven air fuel ratios between the cylinders. As you lean to the Lean of Peak EGT side of the graph, the power starts to drop off. If the air fuel ratio is the same in every cylinder, then every cylinder produces the same amount of power and the engine runs smoothly as you lean. But aviation engine don't always act like that, especially not carburerated flat sixes. They tend to muck up the fuel distribution and some of the cylinders run leaner while others run richer. When this happens each cylinder is producing a different amount of power that results in the engine running rough. Look at the power curve in the best economy band. It varies from 95% down to 85%, but in the best power band the curve it doesn't vary hardly at all. That means if your engine is running rough LOP then as you get richer the engine runs smooth again as the power per cylinder evens out.

This can also happen in flat fours but not to the same degree as flat sixes. Due to larger intake stroke overlap on six cylinder engines the mixture in the intake runners can partially reverse direction under certain conditions, making fuel distribution discrepancies worse. Fuel injected engines can often correct this with GAMI injectors that are adjusted to match their outputs to each cylinder.

Carburerated engines may have a tougher time with this but there are few tricks that may help. First, back off the throttle enough to induce turbulence into the mixture as it passes the throttle plate. This mixes the fuel and air into a more homogeneous mixture before splitting off into the intake runners. If this doesn't work by itself try adding some carb heat to help vaporize the fuel better. These two things may allow the engine to run leaner than before. None of this is guaranteed, but it should help. Try it at a high altitude cruise at about 60% power. You may find you can run leaner than you could before.

Hope this helps.

That is super helpful. Thank you.
It also explains why my engine runs better when leaned and I partially pull carb heat.

 

[MountainDude]

One overlooked factor is spark. Weak sparkplugs will fail to ignite a lean mixture sooner that good plugs, so a plug that is dropping out will create roughness as one cylinder's burn is slower.

How does a spark plug become weak? Is a strong spark plug simply a cleaned/new one, or a specific type?

 

[Martin Pauly]

The real culprit is uneven air fuel ratios between the cylinders. As you lean to the Lean of Peak EGT side of the graph, the power starts to drop off.

To illustrate this, here is a chart from the Mixture Management presentation I give in Oshkosh each year. It's pretty easy to see how two somewhat unbalanced cylinders on the rich side of peak (right side on the chart, cylinders are represented by the green circles) contribute fairly evenly towards generation of horsepower on the shared crankshaft, whereas two cylinders similarly unbalanced on the lean side of peak (left side of the chart) make different horsepower contributions, due to the slope of the HP curve.

 

[Dan Thomas]

How does a spark plug become weak? Is a strong spark plug simply a cleaned/new one, or a specific type?

Every plug has an internal resistor to cut the spark off rather than let it trickle to a stop as the voltage diminishes. This is to reduce electrode erosion. Those resistors have been troublesome in the past, notably in the Champion plugs made up until six or eight years ago. Those had a little resistor pellet that was dropped into the plug, then a small spring on top, and the whole thing held in by a screw. You can see the slot in that screw at the bottom of the connector well. Vibration and heat made that resistor move around, messing up its contacts at either end, and the plugs would fail on the pressure tester after cleaning and gapping. Plugs are supposed to spark to at least 135 PSI. That pressure tester is also supposed to be calibrated so that its voltage, and the gauge's accuracy, are where they are supposed to be.

Auburn was a sparkplug maker that had the best plugs. Their resistor was molded in securely. Those plugs would last many times longer than a Champion. So Champion bought Auburn and shut them down. Tempest and AC started making integral-resistor plugs, and they were nearly as good as the Auburns, but Champion kept insisting that their plugs were just fine. Eventually they finally quietly changed to the Auburn technology they had owned for a decade or more, and their problems seemed to clear up.

This is the reason for resistance checks when cleaning up the plugs. That resistor should have between 1000 and 5000 ohms. Those old Champs would often read infinity, and Champion kept telling us that this was no problem. Yeah, right. Infinity means there's arcing going on in there, and it's not going to get better.

The pressure check will also reveal a cracked ceramic insulator. You'll sometimes hear the crackling of an arc in the well. A contaminated well can do it too, with the spark travelling along the surface to the steel shell at the connector end. The spark at the electrodes will stop while this is happening.

A magneto that hasn't been maintained properly will deliver a weak spark and cause roughness. Most of the time this is due to erosion of the points and/or wear of the cam or rubbing block. This alters the E-gap, the angle of the rotor past neutral when the points open, and the spark generation is weaker.

Too many folks think a spark is a spark. Nope. A weak spark will reduce power, cause bigger mag drops, roughness when lean, and so on. A nice hot spark is what you want. There is no fooling around in that cylinder when a hot spark zaps the mixture.

The ignition harness leads are also a factor here. There is a tester for that.

 

[Silvaire]

As said the roughness is because one or two cylinders are giving up before the rest. If you have a really well balanced set of injectors (this is not going to apply to carbureted engines) with all cylinder, spark plugs and leads in good shape you won't feel any roughness, it will just start losing power.

 

[Capt. Geoffrey Thorpe]

Even with perfectly equal cylinders (which never happens), an engine will start running rough when you get too lean due to the significant increase in cycle to cycle combustion differences. If you ever look at plots of the pressures inside a cylinder, you see changes from cycle to cycle even when the engine is running "smooth".

 

[szwinger]

This is from memory from about 15 yrs ago. The carb'ed O470 reportedly doesn't distribute fuel/air evenly between the sides/banks of cylinders. They make a "crossover balance tube" in two different sizes. When I had my engine P.Ponked (now Northpoint Aviation) the balance wasn't very good and one side "Peaked" before the other side resulting in rough running. It was suggested I try a different balance tube (oddly I think I went smaller but I'm not sure). That made a big difference. It would run very nicely LOP, but I usually flew at 7,500 - 8,500 so generally just ran where ever the engine was smoothest without regard to LOP/ROP. My feeble memory seems to recall it would run at 10.5 GPH which was LOP but I think it ran smoothest around 11~12 GPH.

All of that babbling could be reduced to: Try a different sized crossover balance tube and see if that helps. If not, put the old one back on.

Of course this will only help if all three cylinders on each side peak about the same time (GPH).

 

[MountainDude]

To illustrate this, here is a chart from the Mixture Management presentation I give in Oshkosh each year. It's pretty easy to see how two somewhat unbalanced cylinders on the rich side of peak (right side on the chart, cylinders are represented by the green circles) contribute fairly evenly towards generation of horsepower on the shared crankshaft, whereas two cylinders similarly unbalanced on the lean side of peak (left side of the chart) make different horsepower contributions, due to the slope of the HP curve.

View attachment 132296

Thank you for the graph and explanation. It makes it clearer when presented this way.
I am trying to figure out where is lambda = 1 on the graph. I thought it would be at peak EGT, but the power curve is throwing me off. Why is the power not the highest at peak EGT?

 

[Capt. Geoffrey Thorpe]

I am trying to figure out where is lambda = 1 on the graph.

Purd near at peak EGT

I thought it would be at peak EGT, but the power curve is throwing me off. Why is the power not the highest at peak EGT?

Power is higher rich of lambda=1 - you get preferential burning of the hydrogen in the fuel which gives more energy per O2 molecule. But the unburned fuel left over raises the molecular weight of the cylinder contents and keeps the temperatures down. Also, there is a bit more cooling from evaporating the excess fuel.

 

[William Pete Hodges]

Thank you for the graph and explanation. It makes it clearer when presented this way.
I am trying to figure out where is lambda = 1 on the graph. I thought it would be at peak EGT, but the power curve is throwing me off. Why is the power not the highest at peak EGT?

Part of the reason is a perfect lambda=1 mixture might not completely burn all the oxygen because the air and fuel might not be completely homogenized during combustion. Thus might leave some stray oxygen and fuel molecules left unburned. Then getting slightly richer burns all of the oxygen but also creates a very fast flame propagation causing the pressure to build too close to TDC, thus making excess pressure instead of power to the crank. After that, getting slightly richer slows down flame propagation just enough to put peak pressure about 10 degrees after TDC where the pistion is speeding up and there is enough crank angle for the pressure to produce torque and do useful work.

Variable ignition timing could be set up to improve this, but most aircraft engines have fixed ignition.

 

[Pinecone]

The reason it starts to run rough is that power output is very closely tied to fuel flow/leanness when lean of peak. When rich of peak, the power output does not change much for different fuel flows.

So, if the cylinders are not running at the same leanness, so they are making different levels of power, so rough running.

In the chart below, you can see how the power varies greatly with leanness, when lean of peak. So things like GAMIjectors get all the cylinder running at the same leanness.

 

[MRC01]

... However Continental says, as long as you are at or below the max continuous, you can run anything from full rich to so lean the engine is misfiring without hurting anything.

In the short term, sure. In the long term, the vibration from running rough takes its toll not just on the engine but everything physically connected to it: mounts, alternator, etc.

The reason it starts to run rough is that power output is very closely tied to fuel flow/leanness when lean of peak. When rich of peak, the power output does not change much for different fuel flows.

Exactly. The power vs. mixture curve is asymmetric. It drops off slowly rich of peak, and it drops off quickly lean of peak.

... It also explains why my engine runs better when leaned and I partially pull carb heat.

You can use this in reverse too. Apply carb heat and check the engine response. If you get a noticeable RPM drop then your mixture is rich. If you get little or no RPM change, the mixture is lean (this is under normal conditions, assuming you don't have carb ice). I prefer not to use carb heat except when necessary, because it bypasses the engine air filter.